Garbage throwing personnel tracing method and system based on video images

Abstract

The invention discloses a garbage throwing personnel tracing method and system based on video images, and relates to the technical field of intelligent management and recognition. The system comprises a video acquisition module which is based on a camera of a garbage throwing area and is used for realizing acquisition of a personnel face image and acquisition of a throwing behavior image; the image preprocessing module is used for preprocessing the information acquired by the video acquisition module; the integral system management module is used for carrying out quantitative scoring on different types of garbage throwing behaviors based on the formulated integral rule; and the personnel information storage library is used for storing basic identity information, historical delivery records and integral information of personnel. According to the system, quantitative scoring is carried out on different types of garbage throwing behaviors according to a formulated integral rule, the integral is calculated and updated in real time, and residents are effectively stimulated to correctly throw garbage in a classified mode; and meanwhile, the enthusiasm of residents is further promoted by a point ranking list and a regular point report.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent management and identification, and particularly relates to a garbage throwing personnel tracing method and system based on video images. BACKGROUND

[0002] With the acceleration of urbanization, the urban population is growing, and the production of urban household garbage is also increasing sharply. The traditional manual supervision method is not effective in the face of a large amount of garbage throwing. Manual patrol is not only inefficient and cannot achieve comprehensive coverage, but also cannot accurately trace and effectively supervise the behavior of not classifying and throwing garbage according to regulations. This leads to the phenomenon of random garbage throwing by some residents, which seriously affects the effect of garbage classification and brings many inconveniences to the subsequent garbage disposal work, such as increased garbage disposal cost, waste of recyclable resources, etc., and has a negative impact on urban environmental health and the quality of life of residents.

[0003] Some garbage throwing monitoring systems on the market have many limitations. For example, some monitoring devices can only realize simple video recording function, lack intelligent analysis capability, and cannot automatically identify whether the behavior of garbage throwing personnel conforms to the regulations and determine the identity information. In addition, the existing systems often focus on monitoring the garbage throwing behavior itself, lack effective personnel tracing and long-term tracking management mechanism for illegal behavior, and cannot fundamentally solve the garbage throwing supervision problem.

[0004] According to the search, the application number CN202010131719.4 discloses a garbage classification throwing system and method based on image recognition and integral reward. For household garbage, the garbage classification throwing system includes a classification throwing garbage can, an intelligent composite pole, a garbage classification throwing management system, and a garbage throwing integral management APP. The classification throwing garbage can is used for classified storage of garbage. The intelligent composite pole is provided with an intelligent camera and a prompt device. The intelligent camera is used for collecting and identifying real-time image information of the classification throwing garbage can and the garbage identification area. The deficiency of the application is that it cannot continuously encourage personnel to correctly classify and throw garbage, and lacks effective personnel tracing and long-term tracking management mechanism for illegal behavior. SUMMARY

[0005] The present application relates to the technical field of intelligent management and identification, and particularly relates to a garbage throwing personnel tracing method and system based on video images.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions: A garbage throwing personnel tracing system based on video images, comprising: The video acquisition module is based on a camera in the garbage disposal area, and realizes acquisition of a personnel face image and acquisition of a disposal behavior image. The image preprocessing module pre-processes information acquired by the video acquisition module. The integral system management module quantifies and scores different types of garbage disposal behaviors based on the established integral rules. The personnel information repository stores basic identity information, historical disposal records, and integral information of the personnel. The voice broadcast module customizes personalized voice broadcast content according to the integral information of the personnel and the type of garbage currently disposed. The SMS notification module formulates an SMS content template according to different notification scenarios and purposes, and sends it to the mobile phone of the personnel pre-registered. The timing analysis module analyzes the disposal of the personnel based on a preset time period.

[0007] Preferably, the camera of the video acquisition module includes a personnel recognition camera, an image acquisition camera, and a garbage recognition camera. The garbage disposal area is provided with a garbage room. The image acquisition camera is arranged outside the garbage room. A plurality of electrically controlled doors are arranged on the garbage room. The personnel recognition camera is arranged on the electrically controlled door. A plurality of garbage cans are arranged in the garbage room. The garbage cans are matched with the positions of the electrically controlled doors one by one. The garbage recognition camera is arranged in the garbage room and above the garbage cans. When the personnel enters the garbage disposal area, the video information is acquired based on the image acquisition camera, the personnel face information is acquired based on the personnel recognition camera, and the garbage information is acquired based on the garbage recognition camera.

[0008] Preferably, the image preprocessing module specifically includes the following steps. An image denoising algorithm is used to process the original image to remove noise interference. A histogram equalization algorithm is used to improve the contrast of the image, so that the gray scale distribution of the image is more uniform. An adaptive platform histogram equalization algorithm is used to adjust the contrast of the local area. A face detection algorithm based on deep learning is used to quickly and accurately detect the face area of the personnel in the preprocessed image, and the detected face image is standardized.

[0009] Preferably, the image preprocessing module uses a histogram equalization algorithm to improve the contrast of the image, specifically as follows. First, calculate the gray scale histogram of the image , wherein represents the gray scale level:

[0010] , wherein and are the number of rows and columns of the image, respectively; is an indicator function that is equal to 1 when the gray value of the image at position is equal to , and 0 otherwise. .

[0011] The cumulative distribution function is then calculated as:

[0012] The equalized gray value is finally calculated as:

[0013] where is the total number of gray levels; and is the rounding function.

[0014] Preferably, the integral system management module specifically: quantifies and scores different types of garbage disposal behaviors based on established integral rules; gives corresponding integral rewards or penalties according to the time and frequency factors of garbage disposal; the system calculates the integral of the personnel in real time according to the video acquisition and analysis results; after a disposal behavior is completed, the system immediately calculates the corresponding integral change value according to the type of garbage disposed this time and the classification accuracy, and updates the integral table in the personnel information storage library; based on the integral ranking list, the information of personnel with high integral ranking is displayed to encourage residents to actively participate in correct disposal; at the same time, the integral situation of personnel is regularly statistically analyzed to generate an integral report.

[0015] Preferably, the integral calculation of the integral system management module is specifically as follows: assuming there are n types of garbage, each type of garbage corresponds to a basic score value ; when a user disposes a type of garbage, if the disposal is correct, the integral obtained by the user is calculated by the following formula:

[0016] wherein is a basic weight coefficient, representing the integral weight obtained by correct disposal of garbage; is an additional reward weight coefficient; if the user disposes incorrectly, the integral to be deducted is: ​​

[0017] wherein, is the error deposit penalty weight coefficient.

[0018] The integral update formula is as follows: Let the initial integral of the user be , the new integral after a deposit behavior is:

[0019] If the user deposits correctly this time, then ; if the deposit is wrong, then .

[0020] Preferably, the personnel information repository, in particular: Adopting a combination of a relational database and a distributed file storage system architecture; the relational database is used to store structured data, and the distributed file storage system is used to store unstructured data; Establishing a database table structure, including a personnel basic information table, a deposit record table, and an integral table; When a new personnel first deposits garbage, their basic information is entered into the personnel information repository through manual registration or self-service registration; at the same time, the system automatically collects the facial feature information of the personnel and stores it in association with the basic information; each time the personnel deposits garbage, the system updates their deposit record according to the recognition result.

[0021] Preferably, the voice broadcast module, in particular: According to the historical scoring situation of the personnel and the type of the currently deposited garbage, customized personalized voice broadcast content is played; for personnel with a high historical score, encouraging and prompting voice information is played; for personnel with a low historical score or who frequently violate the rules, guiding voice information is played; For different types of garbage, corresponding voice prompt content is designed, including the correct way of garbage deposit and matters needing attention; based on the opening of the corresponding electric control door of the garbage room, corresponding voice prompt content is played.

[0022] Preferably, the timing analysis module, in particular: Based on the set timing analysis time period, the system automatically starts the analysis program to comprehensively analyze the personnel deposit situation of the day or the week; Data analysis tools and techniques are used to process and analyze the data in the personnel information repository, and the analysis results are displayed to the management personnel and relevant staff.

[0023] A garbage deposit personnel tracing method based on video images, realized based on the above system, comprising the following steps: S1: The personnel enter the garbage deposit area and deposit garbage; S2: The video acquisition module realizes the collection of personnel face images and the collection of the image of the garbage disposal behavior based on the camera of the garbage disposal area; S3: The image preprocessing module pre-processes the information collected by the video acquisition module; S4: The integral system management module quantifies and scores different types of garbage disposal behaviors based on the formulated integral rules; S5: The voice broadcast module plays the customized personalized voice broadcast content according to the integral information of the personnel and the current garbage type; S6: The short message notification module sends the corresponding short message to the mobile phone pre-registered by the personnel; S7: The timing analysis module analyzes the disposal situation of the personnel based on the preset time period, and updates the personnel information repository.

[0024] The beneficial effects of the present application are: The system of the present application quantifies and scores different types of garbage disposal behaviors according to the formulated integral rules, calculates and updates the integral in real time, effectively encourages residents to correctly classify and dispose of garbage; at the same time, the integral ranking list and the regular integral report further promote the enthusiasm of the residents.

[0025] According to the historical scoring situation of the personnel and the current disposal behavior, the system provides personalized voice feedback to help the residents understand the correctness and improvement direction of their disposal behavior, and enhances the interactivity and user experience of the system.

[0026] The system of the present application not only monitors and records data in real time, but also automatically starts the analysis program within the set time period, deeply analyzes the disposal situation of the personnel, and displays the results to the management personnel and relevant staff, which helps to optimize management and decision-making.

[0027] Through long-term tracking management and effective incentive mechanism for illegal behavior, the system of the present application helps to improve the environmental awareness of the residents, reduce the phenomenon of random garbage disposal, and thus improve the urban environmental health and the quality of life of the residents.

[0028] The automatic and intelligent monitoring system of the present application reduces the demand for manual patrol, reduces the supervision cost, and improves the supervision efficiency and accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 A flowchart of a garbage disposal personnel tracing method based on video images is proposed for the present application. DETAILED DESCRIPTION

[0030] The technical solutions of the present application will be further described in detail below in combination with specific embodiments.

[0031] Embodiment 1 A garbage disposal personnel tracing system based on video images, comprising: A video acquisition module based on a camera in a garbage disposal area, which realizes the acquisition of personnel facial images and the acquisition of disposal behavior images; An image preprocessing module for preprocessing the information collected by the video acquisition module; An integral system management module for quantitatively scoring different types of garbage disposal behaviors based on the established integral rules; A personnel information repository for storing personnel's basic identity information, historical disposal records, integral information, and other structured data; A voice broadcast module for customizing personalized voice broadcast content according to personnel's integral information and the type of garbage currently disposed; A short message notification module for formulating short message content templates according to different notification scenarios and purposes and sending them to personnel's pre-registered mobile phones; A timing analysis module for analyzing personnel's disposal accuracy, the number of illegal disposal, the disposal proportion of different types of garbage, and other indicators based on a preset time period, for updating the personnel information repository.

[0032] Among them, the camera of the video acquisition module includes a personnel recognition camera, an image acquisition camera, and a garbage recognition camera. A garbage room is set up in the garbage disposal area. The image acquisition camera is set outside the garbage room. Multiple electrically controlled doors are set up on the garbage room. The personnel recognition camera is set on the electrically controlled door. Multiple garbage cans are set up inside the garbage room, and the positions of the garbage cans and the electrically controlled doors are matched one by one. The garbage recognition camera is set inside the garbage room and above the garbage cans. When personnel enter the garbage disposal area, video information is collected based on the image acquisition camera, personnel facial information is collected based on the personnel recognition camera, and garbage information is collected based on the garbage recognition camera.

[0033] Among them, the image preprocessing module specifically: An image denoising algorithm (such as a wavelet transform-based denoising algorithm or a non-local mean denoising algorithm) is used to process the original image and remove noise interference. A histogram equalization algorithm is used to improve the contrast of the image and make the gray scale distribution of the image more uniform. An adaptive platform histogram equalization algorithm is used to adjust the contrast of local areas. A face detection algorithm based on deep learning is used to quickly and accurately detect the face area of personnel in the preprocessed image, and the detected face image is standardized.

[0034] Among them, the image preprocessing module uses a histogram equalization algorithm to improve the contrast of the image, specifically as follows: First, calculate the gray histogram of the image where, represents the gray level (value range 0-255):

[0035] where, and are the number of rows and columns of the image, respectively; is an indicator function, when the gray value of the image at position is , , otherwise .

[0036] Then the cumulative distribution function is calculated:

[0037] Finally, the equalized gray value is calculated according to the following formula:

[0038] where, is the total number of gray levels (for an 8-bit gray image, ); represents the rounding function.

[0039] The integral system management module, specifically: Based on the established integral rules, different types of garbage disposal behaviors are quantified and scored; for example, correctly classified and disposed of garbage can obtain a certain number of points (such as [X] points), and incorrectly classified and disposed of garbage will be deducted the corresponding points (such as [-X] points); for personnel who repeatedly violate the rules and refuse to correct, an additional integral punishment mechanism is set (such as deducting [X] points at one time), and they are listed in the key supervision list; According to factors such as the time of garbage disposal (such as giving points for disposing of garbage within the specified disposal time), frequency (such as giving additional points to personnel who frequently dispose of garbage and correctly classify them), etc., give corresponding integral rewards or punishments; The system calculates the personnel's points in real time according to the video acquisition and analysis results; after each disposal behavior, the system calculates the corresponding integral change value according to the type of garbage disposed of, the accuracy of classification, etc. according to the integral rules, and updates it to the integral table in the personnel information storage library; Based on the integral ranking list, the information of personnel with high ranking is displayed to encourage residents to actively participate in correct disposal; at the same time, the integral situation of personnel is statistically analyzed regularly to generate an integral report, providing data support for subsequent management and decision-making.

[0040] ​The integral system management module is calculated as follows: Assume that there are n types of garbage, and each type of garbage corresponds to a basic score , for example, the score for correctly depositing recyclable materials is , the score for correctly depositing hazardous waste is , etc. When a user deposits a type of garbage, if the deposit is correct, the points obtained can be calculated by the following formula:

[0041] wherein, is the basic weight coefficient, indicating the integral weight obtained by the correct deposit of garbage; is an additional reward weight coefficient, used to encourage users to actively participate in garbage classification (such as depositing within a certain time period, continuously depositing multiple times correctly, etc.); If the user deposits incorrectly, the points deducted are:

[0042] wherein, is the penalty weight coefficient for incorrect deposit.

[0043] The integral update formula is as follows: Let the initial integral of the user be , and the new integral after a deposit behavior is:

[0044] If the user's current deposit is correct, then ; if the deposit is incorrect, then .

[0045] The personnel information repository is specific in that: The architecture combines a relational database (such as MySQL or SQL Server) with a distributed file storage system (such as HDFS or Ceph); the relational database is used to store the basic identity information, historical deposit records, integral information, and other structured data of personnel, facilitating efficient querying and management; the distributed file storage system is used to store feature images, video clips, and other unstructured data of personnel; The database table structure is established, including a personnel basic information table (containing information such as name, ID number, mobile number, and address), a deposit record table (recording information such as time, location, garbage type, and deposit behavior evaluation for each deposit), an integral table (recording the integral changes of personnel), etc.

[0046] The personnel information repository is specific in that: When a new person first disposes of garbage, their basic information is entered into the personnel information repository through manual registration or self-registration. At the same time, the system automatically collects the person's facial feature information and stores it in association with the basic information. Each time the person disposes of garbage, the system updates their disposal record based on the recognition result, including disposal time, garbage type, disposal behavior score, and other information.

[0047] The voice broadcast module, in particular: According to the person's historical scoring and the type of garbage currently disposed of, customized voice broadcast content is played. For people with high historical scores, encouraging and prompting voice information is played (such as "You have always been a role model in garbage classification, please continue to maintain it! This time you are disposing of kitchen waste, please check if the classification is accurate;"). For people with low historical scores or who frequently violate the rules, more detailed guidance voice information is played (such as "You have some problems with your previous garbage classification, this time you are disposing of recyclable garbage, please carefully confirm before disposing of it;"). For different types of garbage (such as recyclables, hazardous waste, kitchen waste, and other waste), corresponding voice prompt content is designed, including the correct way to dispose of garbage, precautions, etc. Based on the opening of the corresponding electric control door of the garbage room (which can be detected by a sensor), the corresponding voice prompt content is played.

[0048] The SMS notification module, in particular: Integrate an SMS sending platform (such as Aliyun SMS service, Tencent cloud SMS service, etc.) to realize the SMS sending function through an API interface; According to different notification scenarios and purposes, develop multiple SMS content templates. For example, when a person's score is poor, send a reminder SMS (such as "Dear user, your recent garbage classification disposal behavior has some problems, please improve it as soon as possible. This time the score is: [X] points."). When a person successfully disposes of garbage, send a confirmation SMS (such as "Dear user, you have successfully disposed of garbage, this time the classification is accurate, thank you for your cooperation! Your points have been updated to [X] points.") According to the person's point change, send a point report SMS regularly (such as sending a summary of the previous month's points on [X] day of each month).

[0049] The timing analysis module, in particular: Based on the set timing analysis time period (such as [X] o'clock in the morning every day, [X] o'clock in the morning every Monday, etc.), determine the appropriate analysis time point according to actual needs and management requirements. At this time point, the system automatically starts the analysis program and conducts a comprehensive analysis of the personnel disposal situation for the day or the week; The data in the personnel information repository is processed and analyzed by using data analysis tools and techniques (such as SQL statement query, data mining algorithm, etc.), and the analysis results are displayed to the management personnel and relevant staff.

[0050] Embodiment 2: A garbage throwing personnel tracing method based on video images, realized based on the system of embodiment 1, comprising the following steps: S1: personnel enter the garbage throwing area and throw garbage; S2: the video acquisition module realizes the acquisition of personnel face images and the acquisition of throwing behavior images based on the camera in the garbage throwing area; S3: the image preprocessing module pre-processes the information collected by the video acquisition module; S4: the integral system management module quantifies and scores different types of garbage throwing behaviors based on the formulated integral rules; S5: the voice broadcast module plays customized personalized voice broadcast content according to the integral information of the personnel and the type of the currently thrown garbage; S6: the SMS notification module sends corresponding SMS to the mobile phone pre-registered by the personnel; S7: the timing analysis module analyzes the throwing situation of the personnel based on the preset time period, and updates the personnel information repository.

[0051] The above is only the preferred specific implementation of the present application, but the protection scope of the present application is not limited to this, any skilled person in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A garbage disposal personnel tracing system based on video images, characterized in that: include: The video acquisition module uses cameras in the garbage disposal area to capture facial images of people and images of garbage disposal behaviors; An image preprocessing module preprocesses the information collected by the video acquisition module; The points system management module quantifies and scores different types of garbage disposal behaviors based on established points rules; Personnel information repository, which stores personnel's basic identity information, historical delivery records, and points information; The voice broadcast module customizes personalized voice broadcast content based on the personnel's score information and the type of garbage currently placed; SMS notification module, which formulates SMS content templates based on different notification scenarios and purposes and sends them to pre-registered mobile phones of personnel; The timing analysis module analyzes the personnel's delivery situation based on the preset time period.

2. The garbage disposal personnel tracing system based on video images according to claim 1 is characterized in that: The cameras of the video acquisition module include a personnel recognition camera, an image acquisition camera, and a garbage recognition camera. A garbage room is set up in the garbage disposal area. The image acquisition camera is set up outside the garbage room. A plurality of electric-controlled doors are set up on the garbage room. The personnel recognition camera is set up on the electric-controlled door. A plurality of garbage bins are set up in the garbage room. The positions of the garbage bins and the electric-controlled doors are matched one by one. The garbage recognition camera is set up in the garbage room and is located above the garbage bin. When people enter the garbage disposal area, the image acquisition camera collects video information, the personnel recognition camera collects facial information of the people, and the garbage recognition camera collects information about the garbage thrown into the area.

3. The garbage disposal personnel tracing system based on video images according to claim 1 is characterized in that: The image preprocessing module is specifically: The image denoising algorithm is used to process the original image to remove noise interference; the histogram equalization algorithm is used to improve the contrast of the image and make the grayscale distribution of the image more uniform; the adaptive platform histogram equalization algorithm is used to adjust the contrast of the local area; Using a deep learning-based facial detection algorithm, the facial area of ​​a person is quickly and accurately detected in the preprocessed image, and the detected facial image is standardized.

4. The garbage disposal personnel tracing system based on video images according to claim 3 is characterized in that: The image preprocessing module uses a histogram equalization algorithm to improve the contrast of the image, as follows: First calculate the grayscale histogram of the image ,in, Represents grayscale: ； in, and are the number of rows and columns of the image, respectively; is an indicator function, when the image is at position The gray value at hour, ,otherwise ; Then calculate the cumulative distribution function : ； Finally, the grayscale value after equalization is calculated according to the following formula : ； in, is the total number of gray levels; Represents the rounding function.

5. The garbage disposal personnel tracing system based on video images according to claim 1 is characterized in that: The points system management module is specifically: Based on the established scoring rules, different types of garbage disposal behaviors are quantitatively scored; Based on the time and frequency of garbage disposal, corresponding points rewards or penalties will be given; The system calculates the personnel's points in real time based on the video collection and analysis results. When a disposal is completed, the system immediately calculates the corresponding point change value according to the type of garbage disposed and the accuracy of classification according to the points rules, and updates it to the points table in the personnel information repository. Based on the points ranking list, the information of people with high points ranking is displayed to encourage residents to actively participate in the correct placement; at the same time, statistical analysis of people's points is conducted regularly to generate points reports.

6. The garbage disposal personnel tracing system based on video images according to claim 5 is characterized in that: The points calculation of the points system management module is as follows: There are n types of garbage, each type of garbage corresponds to a basic score When a user places a type of garbage, if the placement is correct, the points he gets Calculated by the following formula: ； in, is the basic weight coefficient, which represents the weight of points obtained by correctly disposing garbage; is the additional reward weight coefficient; If the user makes a mistake in placing an order, the points will be deducted. for: ； in, is the penalty weight coefficient for incorrect delivery; The integral update formula is as follows: Assume the user's initial points are , the new points after a delivery behavior for: ； If the user's delivery is correct, ; If the delivery error occurs, .

7. The garbage disposal personnel tracing system based on video images according to claim 6 is characterized in that: The personnel information repository specifically includes: It uses an architecture that combines a relational database with a distributed file storage system; the relational database is used to store structured data, and the distributed file storage system is used to store unstructured data; Establish database table structure, including personnel basic information table, delivery record table, and points table; When a new person disposes of garbage for the first time, his / her basic information is entered into the personnel information repository through manual registration or self-registration; at the same time, the system automatically collects the facial feature information of the person and stores it in association with the basic information; each time the person disposes of garbage, the system updates his / her disposal record based on the recognition results.

8. The garbage disposal personnel tracing system based on video images according to claim 1 is characterized in that: The voice broadcast module is specifically: Customize voice announcements based on the user's historical scores and the type of garbage currently being placed. For users with high historical scores, play encouraging and prompting voice messages; for users with low historical scores or those who frequently place illegal garbage, play guiding voice messages. For different types of garbage, corresponding voice prompts are designed, including the correct way to dispose of garbage and precautions; based on the opening of the corresponding electric-controlled door of the garbage room, the corresponding voice prompts are played.

9. The garbage disposal personnel tracing system based on video images according to claim 1 is characterized in that: The timing analysis module is specifically: Based on the set time period for scheduled analysis, the system automatically starts the analysis program to conduct a comprehensive analysis of the personnel deployment situation for the day or week; Use data analysis tools and techniques to process and analyze data in the personnel information repository, and present the analysis results to managers and relevant staff.

10. A method for tracing the source of garbage disposal personnel based on video images, characterized in that: The implementation of the garbage disposal personnel tracing system based on video images according to any one of claims 1 to 9 comprises the following steps: S1: People enter the garbage disposal area and place garbage; S2: The video acquisition module uses cameras in the garbage disposal area to capture facial images of people and images of their garbage disposal behavior; S3: The image preprocessing module preprocesses the information collected by the video acquisition module; S4: The points system management module quantifies and scores different types of garbage disposal behaviors based on the established points rules; S5: The voice broadcast module plays customized personalized voice broadcast content based on the personnel's score information and the type of garbage currently placed; S6: The SMS notification module sends the corresponding SMS to the pre-registered mobile phone of the personnel; S7: The timing analysis module analyzes the personnel deployment situation based on a preset time period and updates the personnel information repository.

Images